We hypothesize that prostate cancer (CaP) associated gene expression signatures in specific cell types of the human prostate gland will define the pathobiology of epithelial and stromal components in prostate tumorigenesis, and a subset of these genes may predict the aggressive form of the disease. The objective of this proposal is to perform comparative genomics analyses on laser capture micro-dissected (LCM)- epithelial and stromal cells from prostate glands of carefully selected CaP patients and define gene clusters predictive of non-aggressive or aggressive form of the disease. Our experience with the gene expression analysis of LCM-epithelial cell RNAs, promising preliminary GeneChip data from LCM-epithelial cell RNAs, and development of a LCM DNA/RNA bank from CaP patients provide the basis for following specific aims proposed here. Aim 1: Identify CaP associated epithelial cells specific gene expression patterns: GeneChip defined expression patterns in paired normal and cancer epithelial cells from two CaP patient groups, one with "aggressive CaP" (PSA recurrence, Gleason score 8-9, T3c stage, seminal vesicle invasion, poor tumor differentiation) and the other with "non-aggressive CAP" (no PSA recurrence, Gleason score 6-7, T2a-T3b stage, no seminal vesicle invasion, well or moderate tumor differentiation), will be analyzed for CaP specific expression signatures. Emphasis will be placed on the gene clusters that define "aggressive CAP". Aim 2: Identify CaP associated stromal cells specific gene expression patterns: Using similar strategy and same cohort of patients as for Aim1, CaP associated stromal cells specific gene expression patterns will be defined and emphasis will be placed on gene clusters associated with aggressive CaP. Aim 3: Validation of "Aggressive CAP" associated epithelial or stromal cells specific genes: A panel of "aggressive CAP" associated genes will be analyzed in a large cohort of patients by tissue microarray and/or quantitative real time PCR assay and its predictive value will be established. Significance: Epithelial or stromal cells specific gene expression patterns characteristics of CaP with potential roles in pathobiology will be defined. Gene clusters predictive of aggressive CaP has potential in: (1) providing novel modalities in the future clinical management of CaP; and (2) in identification of novel therapeutic targets for aggressive CaP.